Bottom-up computer simulations (as they are used, for example,
in Artificial Life) are changing the way we explore dynamical
systems, such as cities and the societies they house. With
a short historical view, introducing the distinction between
capital and metropolis (say, Rome and Venice in the Middle
Ages, or Washington and New York today), I will explain
how some historians (Braudel, McNeill) now view the rapid
rise of the West as partly attributable to the dynamics
of metropolises. From here we will get into a philosophical
discussion of the issues raised in the conference, as they
can be explored via virtual environments within computers.
Understanding how the virtual worlds created by computer
simulations can allow us to go beyond the limitations of
top-down analytical techniques (and to instead "synthesise
a dynamical system" from the bottom-up) will involve a discussion
of the concepts from a variety of disciplines: The far-from-equilibrium
dynamics of Prigogine, the non-linear conceptsof chaos and
complexity theory, and the "population thinking" which characterises
neo-Darwinism.
The discipline of Artificial Life (AL),
in which evolutionary questions are explored by unleashing
a population of virtual animals within a computer, and the
evolution of the population over many generations explored,
is a good example of a modern research program which embodies
ideas from all three fields above. However, the exact same
techniques used in AL can be adapted for the study of urban
dynamics.
